The purpose of this paper was to compare two methods of fractionation of reactive phosphorus forms (RP) in bottom sediments of polymictic water reservoirs located in ruralareas. As a criterion for estimation, quantities RP extracted in analyzed fractions versusthe general content of Al, Fe and Ca were used. The sediments were collected in springtime, from rural areas with different land use in the drainage basin (arable fields, uncultivatedland, farmyards). The sediments were collected with a KC-Denmark core samplerfor bottom sediments, which makes it possible to collect samples leaving the upper structureof 6 cm in thickness intact. After mineralization, each sample was analyzed for the general phosphorus content and a speciation analysis was performed by two methods. Withthe first method (M1), the RP bonded to Al (Al_RP), Fe (Fe_RP) and Ca (Ca_RP) was determined. The second method (M2) was applied to determine fractions of reactive phosphorus released from the sediments under reduced conditions, mainly from bonds withiron and manganese (BD_RP), adsorbed on aluminium oxides mainly Al2O3 and other surfaces (NaOH_RP) and apatite phosphorus and bonds with carbonates. In almost all the sediments, most RP was determined in the phosphorus fractions with Ca extracted with acid solution. For the fractions obtained via the M1 method, more of the general RP, as well as phosphorus bonded to calcium and iron was found than for fractions obtained with the M2 method, and the differences might be attributed to translocation of phosphorus from organic to inorganic compounds during fractioning. The fractionof reactive phosphorus determined with the M1 method is correlated more strongly withthe general content of the metal in the sediments than reactive phosphorus fractions determined with the M2 method. The strong correlation between the content of iron and the fraction of BD_RP in mineral sediments shows that it can be an indicator of the sensitivity of sediment to changes in the redox potential.